Application of a multifractal analysis to study brain white matter abnormalities of schizophrenia on T2-weighted magnetic resonance imaging

Multifractal analysis provides a precise quantitative description of the structural complexity of white matter (WM) on magnetic resonance imaging (MRI). To test this new technique as an aid to elucidating the pathology of schizophrenia, we examined a multifractal dimension (i.e. Δα) of WM in schizophrenia patients and their relations to clinical variables. We examined 16 patients with schizophrenia and 16 controls matched for age, sex and handedness. Delta α value of WM in the prefrontal and frontoparietal lobes and the corpus callosum (genu and splenium) on T2-weighted MRI was calculated. Delta α was not significantly different between groups in either region of interest. However, group-by-side interaction for Δα was found in the frontoparietal WM; post-hoc analysis revealed normal left dominant asymmetry in Δα for frontoparietal WM in control subjects, which was absent in schizophrenia patients. Furthermore, the patients with schizophrenia had a lower asymmetry coefficient ([R − L]/[R + L]) for Δα in frontoparietal WM. Relations to clinical symptoms from the Positive and Negative Syndrome Scale, Δα in corpus callosum, and the asymmetry coefficient in prefrontal WM were correlated with negative and general psychopathology symptom scores. Our results support the left-sided dysfunction hypothesis of schizophrenia and its relation to schizophrenic symptoms. Multifractal analysis reveals abnormal patterns of WM structures in schizophrenia that could be implicated in the disorder's etiology.